RADIATE APPEARANCES DUE TO DIFFUSION 249 



menon which possesses a certain similarity with the forma- 

 tion of these arese. It is well known that round rapidly 

 growing crystals of coloured substance which are separating 

 .out from a saturated or supersaturated solution, there 

 frequently arises a distinct area which is coloured slightly 

 or not at all. This is explained by the fact, that the 

 rapidly growing crystal attracts the dissolved substance of 

 the surrounding fluid more quickly than it can travel by 

 diffusion from the surrounding fluid. The formation of 

 such an area is, for instance, always to be observed when a 

 crystal grows in a saturated solution by means of so-called 

 globulites, i.e. minute drops of strongly supersaturated solu- 

 tion. Then there is always found round the nucleus a cleUr 

 area free from such globulites, which depends on the fact 

 that the globulites are gradually dissolved in the solution 

 already rendered dilute by the formation of the crystal 

 (see on this point Lehmann, Molecularpliysik, Bd. i. pp. 319 

 and 726 et seq.). Now in a similar manner to the grow- 

 ing crystal the central body draws to itself substances from 

 its surroundings, and if this attraction goes on rapidly, it 

 may happen, as in the growth of the crystal, that the diffu- 

 sion from the surroundings does not furnish a supply with 

 sufficient rapidity, in consequence of which the relative 

 degree of concentration is disturbed in a certain area, and 

 in this region chemical alterations can be effected. Whether 

 these alterations, as in the case of the globulites, may 

 eventually manifest themselves by the solution of granules 

 and their attraction to the central body, I leave an open 

 question. This certainly does not seem impossible, since the 

 central bodies increase with every division of the cell, and do 

 not diminish in mass, thus undergoing a gradual growth. If 

 we consider, however, that the tingibility of the protoplasmic 

 substance is often very considerably impaired by relatively 

 slight influences I will only cite in this respect the char- 

 acteristic red coloration of the chromatin granules of the 

 nucleus and protoplasm by hsematoxylin which I have 

 described, a reaction which almost always fails after previous 

 fixation with acids it may also be considered possible that 

 the slight, or rather the special tingibility of the central 



